This small SQL snippet splits a String into Rows and Columns and returns a TABLE.
5 years ago, I had developed a Hotel Management Application for one of my clients. Like all projects I did in those days, never really thought about the changes I was doing and the impact they would have on the application in future.
Anyways, recently the Hotel contacted me and asked me to show date wise Stock Update history Vs Items used.
The table I had designed back then was as follows:
|ServiceID ||Service Name ||Stock ||Stock Update History (Stock Added|Date) |
|1 ||TEA ||14 ||4|22/05/13#5|04/06/13 |
|2 ||MINERAL WATER ||50 ||20|21/06/13#15|19/06/13#15|14/06/13 |
As you can see the Stock Update history column was stored a single string wherein line items were seperated by # and columns with a pipe sign. Every time stock was added, I would append #StockNo|Date to the Stock Update History column.
Since I now had to do some math on these values, it was essential to parse these values in a Table.
Using the code
I started by using a split function from SQLCentral.
CREATE FUNCTION [dbo].[Split](@String varchar(8000), @Delimiter char(1))
returns @temptable TABLE (items varchar(8000))
declare @idx int
declare @slice varchar(8000)
select @idx = 1
if len(@String)<1 or @String is null return
while @idx!= 0
set @idx = charindex(@Delimiter,@String)
set @slice = left(@String,@idx - 1)
set @slice = @String
insert into @temptable(Items) values(@slice)
set @String = right(@String,len(@String) - @idx)
if len(@String) = 0 break
To call the function you would simply write:
Select * from Split('20|21/06/13#15|19/06/13#15|14/06/13','#')
And what you would get is:
The function worked great, but it would only split the string into 1 column.
I looked up google and found a function shared by bvr on Stackoverflow which would split string into columns.
I merged these two functions and what I got was a Function that would split strings based on a Line delimiter (#) and a Column delimiter (pipe sign) and output to a Table.
DECLARE @SUH VARCHAR(2000);
SET @SUH=(Select [Stock Update History] from ServicesAvailable where ServiceID=1);
WITH Split_Names (Name, xmlname)
+ REPLACE(TA.items,'|', '</name><name>') + '</name></Names>') AS xmlname
FROM (Select * from Split(@SUH,'#')) TA
xmlname.value('/Names/name','varchar(100)') AS Stock,
xmlname.value('/Names/name','varchar(100)') AS UpdatedOn
Points of Interest
As you can see the code is sourced from a couple of places. The reason I am sharing this is because it is difficult to find a single function that does that. I am sure people out there will have a better way of doing this and I acknowledge the fact that I am a noob when it comes to SQL.
Anyone with a better solution, please do share - we are all learners here (probably on different levels),
First version - no history.